INDEX

Explanations

references to physical stalls or obstacles in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "stall" and its variations

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

vernment

-0.90

 GOODMAN

-0.77

âĵĺ

-0.72

vironment

-0.69

é¾įåĸļå£«

-0.67

éĹĺ

-0.67

perty

-0.65

*/(

-0.65

ilitary

-0.65

arag

-0.65

POSITIVE LOGITS

ions

1.01

 Stall

0.97

ioned

0.96

 stall

0.94

 stalls

0.89

ings

0.88

worth

0.86

ion

0.84

acles

0.84

asse

0.84

Activations Density 0.035%

references to physical stalls or obstacles in various contexts

occurrences of the word "stall" and its variations

No Comments

No Known Activations