INDEX

Explanations

phrases related to conversations or discussions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "talking."

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

emale

-0.79

eele

-0.78

\/\/

-0.77

iverpool

-0.71

quez

-0.70

avorite

-0.70

Insp

-0.69

reci

-0.68

ettel

-0.67

âĢ¢âĢ¢âĢ¢âĢ¢

-0.66

POSITIVE LOGITS

 aloud

0.88

 loudly

0.81

 about

0.77

bone

0.76

 Heads

0.76

 louder

0.75

 amongst

0.70

heads

0.70

ebus

0.69

bags

0.69

Activations Density 0.019%

phrases related to conversations or discussions

occurrences of the word "talking."

No Comments

No Known Activations