INDEX

Explanations

phrases related to specific concepts or objects, potentially with negative connotations

oai_token-act-pair · gpt-3.5-turbo

metaphorical expressions that imply deception, manipulation, or undesirable outcomes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

URA

-0.79

cont

-0.78

rongh

-0.77

alle

-0.76

ãĤ¼ãĤ¦ãĤ¹

-0.73

ickets

-0.72

qus

-0.71

Rail

-0.71

rab

-0.71

arel

-0.70

POSITIVE LOGITS

 mentality

1.04

 scenario

0.99

 approach

0.98

 tactic

0.92

 moment

0.86

 situation

0.85

 solution

0.85

 fallacy

0.84

 maneuver

0.82

 tactics

0.82

Activations Density 0.378%

phrases related to specific concepts or objects, potentially with negative connotations

metaphorical expressions that imply deception, manipulation, or undesirable outcomes

No Comments

No Known Activations