INDEX

Explanations

words related to geographical features, specifically hills

oai_token-act-pair · gpt-3.5-turbo

references to hills

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

uality

-0.88

ECA

-0.72

ãĥ¯

-0.71

âĹ¼

-0.70

 Role

-0.70

~~~~

-0.69

âĸĳâĸĳ

-0.69

Oracle

-0.68

âĸ¬

-0.66

 Attention

-0.66

POSITIVE LOGITS

tops

1.03

 hill

1.01

side

1.01

 hills

0.93

frog

0.92

top

0.91

 slopes

0.91

stead

0.84

 climbs

0.80

keeper

0.79

Activations Density 0.007%

words related to geographical features, specifically hills

references to hills

No Comments

No Known Activations