INDEX

Explanations

words related to conflicts and their impact

oai_token-act-pair · gpt-3.5-turbo

instances of the word "conflict"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

prints

-0.86

icrobial

-0.86

alty

-0.75

Reviewer

-0.73

GC

-0.71

printed

-0.69

FORMATION

-0.69

 shelves

-0.69

arers

-0.68

vard

-0.67

POSITIVE LOGITS

 raging

0.99

 raged

0.86

 Royale

0.86

 resolution

0.85

naire

0.85

fighting

0.82

 Resolution

0.82

 escalation

0.82

 waged

0.81

 unresolved

0.81

Activations Density 0.023%

words related to conflicts and their impact

instances of the word "conflict"

No Comments

No Known Activations