INDEX

Explanations

terms related to physical structures or locations, specifically mentioning the word "fount" with varying levels of emphasis

oai_token-act-pair · gpt-3.5-turbo

references to mountainous or elevated terrains

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

pora

-0.61

RGB

-0.60

nexus

-0.60

 Shattered

-0.60

 Situation

-0.59

Else

-0.59

Gut

-0.59

STD

-0.58

cci

-0.58

gging

-0.57

POSITIVE LOGITS

ount

1.35

iful

1.19

enance

0.97

ifully

0.95

eous

0.95

gomery

0.86

sburg

0.86

ette

0.85

anamo

0.83

icularly

0.83

Activations Density 0.006%

terms related to physical structures or locations, specifically mentioning the word "fount" with varying levels of emphasis

references to mountainous or elevated terrains

No Comments

No Known Activations